# aisix vs agentgateway

Operated by the Busbar team; Busbar is also an entrant. Gateway overhead against a local mock is not end-to-end model latency.  [Methodology and limitations](https://onthebench.ai/gateways/method) .

Requested cell: openai → openai. Metric: added_latency_p99_us.

Comparability: comparable. Ratios require an identical protocol pair, hardware, architecture, metric definition, clean harness commit, recorded mock digest, and metric-specific workload. Missing evidence never matches missing evidence. These are observed gateway measurements against a local mock, not a claim about model latency or statistical significance.

- Load-generator binary digests were not recorded; instrument equivalence is based on the identical clean harness source commit and mock digest.
- Measurements were made at different times on independently provisioned machines.

Ratio (B ÷ A): 1.4425287356321839. No statistical significance is asserted.

[Comparison JSON](https://onthebench.ai/api/v1/compare?a=aisix&b=agentgateway&ingress=openai&egress=openai&metric=added_latency_p99_us)  ·  [All AISIX (api7) evidence](https://onthebench.ai/gateways/aisix)  ·  [All agentgateway evidence](https://onthebench.ai/gateways/agentgateway)

## aisix

Build ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a. Measured 2026-11-10T17:39:23Z.  [Run evidence](https://onthebench.ai/gateways/runs/run-aisix-ef0cff28391e0017f17c4b7c01e0171c62359468b787de48402bc3408df76915)

| Ingress → upstream | Metric | Value | State / reason | p99 bound (ms) |
| --- | --- | --- | --- | --- |
| openai → openai | added_latency_p50_us | 146 | valid; gateway_c1_samples=35048; direct_c1_samples=262627 | — |
| openai → openai | added_latency_p99_us | 174 | valid; gateway_c1_samples=35048; direct_c1_samples=262627 | — |
| openai → openai | gateway_c1_p99_us | 202 | valid; gateway_c1_samples=35048 | — |
| openai → openai | direct_c1_p99_us | 28 | valid; direct_c1_samples=262627 | — |
| openai → openai | cpu_us_per_request | 152.64101169979259 | valid | — |
| openai → openai | rps_per_cpu_second | 6551.319261213721 | valid | — |
| openai → openai | cost_window_conc | 8 | valid | — |
| openai → openai | cost_core_utilisation | 0.9445269209542334 | valid | — |
| openai → openai | cost_window_ok | 148977 | valid | — |
| openai → openai | cost_window_rps | 24826 | valid | — |
| openai → openai | cost_threads | 16 | valid | — |
| openai → openai | cost_nonvol_ctxt_per_request | 0 | valid | — |
| openai → openai | cost_majflt | 0 | valid | — |
| openai → openai | added_ttft_p50_us | 277 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| openai → openai | added_ttft_p99_us | 366 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| openai → openai | added_gap_p50_us | 2 | valid; gap_gw_samples=63; gap_direct_samples=63 | — |
| openai → openai | added_gap_p99_us | below_resolution | below_resolution; gap_gw_samples=63; gap_direct_samples=63 | — |
| openai → openai | streams_sustained_fps | 30259.205544499786 | valid | — |
| openai → openai | streams_sustained | 645 | valid | — |
| openai → openai | idle_rss_mib | 34.2890625 | valid | — |
| openai → openai | steady_state_rss_mib | 68.62890625 | valid | — |
| openai → openai | recovered_rss_mib | 52.296875 | valid | — |
| openai → openai | peak_rss_mib | 70.42578125 | valid | — |
| openai → openai | peak_rss_hwm_mib | 70.28515625 | valid | — |
| openai → openai | time_to_plateau_s | not_measured | not_measured | — |
| openai → openai | growth_rate_mib_per_min | 0.37583951307297037 | valid | — |
| openai → openai | load_s | 300 | valid | — |
| openai → openai | shape | 0 | valid | — |
| openai → openai | idle_shape | -1 | valid | — |
| openai → openai | idle_static | 0 | valid | — |
| openai → openai | idle_growth_rate_mib_per_min | -0.8534572049737088 | valid | — |
| openai → openai | throughput_rps | 28888 req/s | concurrency 16; samples 173346 | 1 |
| openai → openai | throughput_rps | 31374 req/s | concurrency 64; samples 188254 | 5 |
| openai → openai | throughput_rps | 32092 req/s | concurrency 128; samples 192583 | 10 |
| openai → openai | throughput_rps | 32092 req/s | concurrency 128; samples 192583 | 50 |
| openai → openai | throughput_rps | 32092 req/s | concurrency 128; samples 192583 | 100 |
| openai → openai | throughput_rps | 32092 req/s | concurrency 128; samples 192583 | unbounded |

## agentgateway

Build ghcr.io/agentgateway/agentgateway:v1.7.0@sha256:4562330cd6e580bfac2de5b8905fee608dcf3fff6f1f4900744d27bdddb3b41b. Measured 2026-11-10T17:39:19Z.  [Run evidence](https://onthebench.ai/gateways/runs/run-agentgateway-fa351efdcee949e977dd1d4331440512b3e913a74b878f4bd1b19a7c7ca2900c)

| Ingress → upstream | Metric | Value | State / reason | p99 bound (ms) |
| --- | --- | --- | --- | --- |
| openai → openai | added_latency_p50_us | 216 | valid; gateway_c1_samples=24618; direct_c1_samples=233879 | — |
| openai → openai | added_latency_p99_us | 251 | valid; gateway_c1_samples=24618; direct_c1_samples=233879 | — |
| openai → openai | gateway_c1_p99_us | 282 | valid; gateway_c1_samples=24618 | — |
| openai → openai | direct_c1_p99_us | 31 | valid; direct_c1_samples=233879 | — |
| openai → openai | cpu_us_per_request | 188.57584939720198 | valid | — |
| openai → openai | rps_per_cpu_second | 5302.905982905983 | valid | — |
| openai → openai | cost_window_conc | 8 | valid | — |
| openai → openai | cost_core_utilisation | 0.9718871854587116 | valid | — |
| openai → openai | cost_window_ok | 124088 | valid | — |
| openai → openai | cost_window_rps | 20679 | valid | — |
| openai → openai | cost_threads | 13 | valid | — |
| openai → openai | cost_nonvol_ctxt_per_request | 0 | valid | — |
| openai → openai | cost_majflt | 0 | valid | — |
| openai → openai | added_ttft_p50_us | 327 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| openai → openai | added_ttft_p99_us | 360 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| openai → openai | added_gap_p50_us | 10 | valid; gap_gw_samples=63; gap_direct_samples=63 | — |
| openai → openai | added_gap_p99_us | below_resolution | below_resolution; gap_gw_samples=63; gap_direct_samples=63 | — |
| openai → openai | streams_sustained_fps | 32614.06218906021 | valid | — |
| openai → openai | streams_sustained | 2048 | valid | — |
| openai → openai | idle_rss_mib | 34.00390625 | valid | — |
| openai → openai | steady_state_rss_mib | 53.4921875 | valid | — |
| openai → openai | recovered_rss_mib | 49.99609375 | valid | — |
| openai → openai | peak_rss_mib | 54.5234375 | valid | — |
| openai → openai | peak_rss_hwm_mib | 54.2265625 | valid | — |
| openai → openai | time_to_plateau_s | not_measured | not_measured | — |
| openai → openai | growth_rate_mib_per_min | -2.0749110660555545 | valid | — |
| openai → openai | load_s | 300 | valid | — |
| openai → openai | shape | -1 | valid | — |
| openai → openai | idle_shape | not_measured | not_measured | — |
| openai → openai | idle_static | 1 | valid | — |
| openai → openai | idle_growth_rate_mib_per_min | 0.03612607648971414 | valid | — |
| openai → openai | throughput_rps | 21156 req/s | concurrency 8; samples 126952 | 1 |
| openai → openai | throughput_rps | 22239 req/s | concurrency 64; samples 133447 | 5 |
| openai → openai | throughput_rps | 22380 req/s | concurrency 128; samples 134304 | 10 |
| openai → openai | throughput_rps | 22545 req/s | concurrency 256; samples 135283 | 50 |
| openai → openai | throughput_rps | 22545 req/s | concurrency 256; samples 135283 | 100 |
| openai → openai | throughput_rps | 22545 req/s | concurrency 256; samples 135283 | unbounded |
